Slashdot Mirror
Statistical Suspicions In Iran's Election
hoytak writes "An expert in electoral fraud, professor Walter Melbane, has released a detailed analysis (PDF) of available data in Iran's controversial election (summary here). While he did not find significant indications of fraud, he does note that all the deviations from the predicted model are in Ahmadinejad's favor: 'In general, combining the 2005 and 2009 data conveys the impression that a substantial core of the 2009 results reflected natural political process... [These] stand in contrast to the unusual pattern in which all of the notable discrepancies between the support Ahmadinejad actually received and the support the model predicts are always negative. This pattern needs to be explained before one can have confidence that natural election processes were not supplemented with artificial manipulations.'" In related news, EsonLinji notes reports in the Seattle PI and other sources that the US State Department has asked Twitter to delay system maintenance to prevent cutting off Iranians who have been relying on the service during the post-election crisis. And if you would like to help ease the communication crunch, reader RCulpepper tips a blog post detailing how to set up a proxy server for users with Iranian IP addresses.
This isn't electoral interference. It is an attempt to prevent censorship and aid people who are being oppressed and persecuted. This is exactly the sort of intervention that countries should be doing: helping allow more people to talk to each other. Democracy comes most easily not when imposed by a military invasion but when people are simply given the tools necessary to talk to each other and to those from other countries. Dictators always try to censor and control communications for a reason. I'm not that happy with how the Obama administration has done things (especially in regards to civil liberties issues) but this is precisely the correct reaction. Actions that undermine censorship are very rarely the wrong thing.
Meh, The US State Department talking to a US company that provides a services that some Iranians use is hardly a particularly good example of external political influences in the middle east. If anything the big story would be if somebody actually managed to persuade Twitter to keep operating. :) But seriously, when you look at things like Operation Ajax, you can see that the US just trying to make sure Iranians have a convenient way to speak for themselves is extremely hands-off, and probably a very appropriate way to avoid having unclean hands in the situation. The previous administration would have loudly and openly run their mouth about the situation, and inadvertently marginalized the reformist element in Iran by trying to support it. Trying to make sure they can speak for themselves is probably about the best thing America can do right now.
But they didn't just announce that. They even had a claimed final total shortly thereafter which Ayatollah Khamenei confirmed. That's not explainable by a "we have a representative sample".
Except that you have good transit systems and are counting around 30M fewer votes.
1) The counting and reporting is done right from the polling stations. Checks and rechecks, and additional audits might benefit from being able to move the ballots around efficiently, but the initial counting and reporting is very efficient.
2) The number of ballots being counted is completely irrelevant, and would make no difference in how long it takes. We allocate polling stations and staff them on a per capita basis. So if our population doubled it wouldn't take twice as long, we'd just have twice as many polling stations and staff.
(And it would cost twice as much, but there would be twice the population paying for it, so it all works out the same whether its 30M or 300M people.)
The only effort that goes up, is summarizing results, but:
a) that is largely done using computers
b) the amount of effort grows logarithmically so 300M ballots would only require a handful more staff than 30M ballots.
Really, people who think you can't run an efficient paper ballot system with a large population aren't really thinking.
Except that you have good transit systems and are counting around 30M fewer votes.
From what I understand, Iran has a good, or at least decent, transit system, (they aren't a third world country) and a decent communication grid.
Having 4x more votes means nothing. They could easily have 4x (or more) counters.
The US could hand count over 100m ballots in the same time frame, if you only had one election at a time, like Canada. Because you have a FREAKING CRAP TON of elections (President, Senate, House, State Senate, State House, DA, Judge, School Board, Official State Dog Walker) and often several referendums all at one time, hand counting becomes impractical. BTW, I am not bashing having so many elections, just pointing out that it is the major reason why hand counting is impractical in the US.
That's correct. But the opposition candidate, Mousavi, said that he received a phone call at 2am the evening of the election indicating that he had won. When the results were announced later, it was Ahmadinejad by a landslide.
Additionally, A'nejad officially had consistent support all across the country and all through demographics. He officially did equally well in cities vs. rural areas. Mousavi was heavily favored in cities. A'nejad officially did equally well among sexes, age groups, class levels, ethnic groups, everything. Mousavi was heavily favored among young students. It's too uniform to be plausible. For example, A'nejad even beat Mousavi in Mousavi's home Azeri province, Iranian Azerbaijan. That was compared to Obama losing the African-American vote to McCain, it's just very suspect and highly improbable.
In addition to that, the other 2 candidates each officially received less than 1% of the total. In the pre-election polls each of those candidates had much higher support.
CNN has done an absolutely terrible job at covering this, the line that CNN is reporting is essentially the government's spin being reported as truth. Fox seems to be the only US network with the balls to show much protest video. The BBC's coverage has been among the best outside of Arabic media, which is difficult to receive in a lot of places. The most up-to-date information about this can usually be found in whichever fark.com thread people are currently posting in, they've gone through 9 or 10 now with several thousand posts in each. Needless to say, any respect I had for CNN has essentially evaporated. Their international coverage used to be among the best in the US, now they might as well be the US-based Iranian spin machine.
"Our two-party system is like a bowl of shit looking at itself in a mirror." - Lewis Black
It is explainable by a "this is a Theocracy and I am the High Priest."
I don't understand why people act as if they expect Iran to conduct an election like a Western democracy.
Where does the assumption come from that Iran, of all countries, is even capable of a "fair election"? I really don't see how you can be surprised about this outcome *at all*, and I also don't understand what anyone thinks can be "done about it", if Iran's own "government" does not take action.
-fb Everything not expressly forbidden is now mandatory.
>Right. After all, Khamenei's vote is "some potion of the votes."
It would be "legitimate", in the sense that it would be consistent with Iran's law, for Khamenei to simply appoint the President setting aside any other considerations (such as elections.)
The Ayatollah's word is absolute law, constrained only by natural consequences -- say, if the protests grow to the point where they represent an actual rebellion.
-fb Everything not expressly forbidden is now mandatory.
Thank you for such an informative and helpful post, obviously you've been contributing to this cause long before it hit /. We're always griping about threats to free speech and fair elections, but here's something a person with some technical skill can do to combat it.
Also, this has nothing to do with the election results. Even if Ahmadinejad received more votes, silencing the opposition is a major injustice. The fact that everyday joes can thwart his efforts with a PC and internet connection is pretty amazing. The power of the Internet has been subject to a lot of hyperbole and BS, but this is an example of how it really does change history.
The Iranians apparently thought they deserved a fair election. This is not a tempest in a teapot, imagined up by the western world. Watch the videos, read the live feeds through twitter, listen to the chants of "Allaho Akbar" that shake the cities: it is very clear that Iran's leadership dramatically overstepped in this election.
It doesn't matter one single bit that the country is an effective dictatorship. The people were promised an election to choose their own president, and no sooner had they made their choice than the government yanked the promise away from them. It doesn't even matter if a fair counting of every vote cast does indicate a win for Ahmadinejad; the blatant fraud, police brutality, and the arresting of the opposition has ruined the people's trust in government. I truly hope that Iran doesn't descend into civil war.
Just another "DOJ fascist authoritarian totalitarian bootlicker" -- Zeio
So what if they did manipulate the election. What is anybody going to do about it? What has anybody done about the last dozen suspicious elections around the globe?
This is a pretty bad post even by slashdot standards.
You start out by proclaiming "As a PoliSci student, I've spent a ton of time looking at election data for many countries", which as an argument to authority is trying to show how your opinion counts more than other posters (it does not).
This is then followed up by a false analogy "I'm pretty sure some people protested after Kerry lost the 2004 US election". The vast differences between these elections renders the analogy meaningless. Never the less you decide to throw in an ad hominem "and yeah, I know I'll get some conspiracy nut reply to that with an essay" just to reinforce it.
Time for some red herrings:
"Quebec, in particular, has a history of some pretty wild swings from one party to another."
"I've seen countless "rules" made in politics, only to be broken"
"I saw similar analysis "prove" Kerry really won in 2004, and that didn't really amount to anything."
All divert attention towards other barely related topics.
You end by stating that you are "only suggesting that I've not seen any "smoking gun" ", which places an unfair burden of proof upon the opposition. The incumbent (Ahmadinejad) controlled every step of the elections, the smoking gun you are looking for is just not possible with this level of control.
I apologise for pointing out the logical fallicies because usually posts like this annoy me in that they don't address (and therefore dismiss) the arguments but the post had too many problems to ignore.
You mean aside from the bizarre statistical anomalies? I'm starting to get a distinct "la la la I can't hear you"
You're absolutely correct, once I read "did not find significant indications of fraud" I cannot hear anyone talking about statistical anomalies. Well no, not absolutely correct, there's no "la la la" about it. It's just the same cold uncompromising "I can't hear you" a Judge will give if you attempt to submit irrelevant evidence.
You clearly either haven't read the stat breakdown on the official results, don't understand what they mean, or are a lvl 23 Erudite Troll.
Hey FU buddy, I'm at least a lvl 30! ;) You're probably correct about my lack of understanding, so maybe you can help me with this. Quoting from TFA:
... a statistically sharp approach to statistical testing--taking the multiple testing into account--fails to provide evidence against the hypothesis that the second digits are distributed according to Benford's Law. Tests based on the means of the second digits also fail to suggest any deviation from the second-digit Benford's Law distribution ...
I've taken on board the fact that hetrogeneity of the data might obscure local problems, but that's not really relevant since the onus of proof clearly lies with those alleging fraud anyway. I've also taken on board that the non-sharp approach (ie looking at one of the candidates results) might lead one to another conclusion. As to the other (non-significant) anomalities in his modified conclusion Mebane observes:
In general, combining the 2005 and 2009 data conveys the impression that a substantial core of the 2009 results reflected natural political processes. ... These natural aspects of the election results stand in contrast to the unusual pattern in which most of the notable discrepancies between the support Ahmadinejad actually received and the support the model predicts are negative. ... It appears that the specification using the two conditioning variables ... does not fully capture the baseline support for the candidates or the pattern of new mobilization. ... It is not possible given only the current data to say whether this reflects natural complexity in the political processes or artificial manipulations.
I admit I lack the confidence with stats (1978 was the last time I had to look at stats in any deeper way than merely applying tests of significance), to make a call on non-significant results. I get nervous when statisticians talk about "impressions" in place of "significance." Perhaps you can educate me here and convince me why I should draw conclusions based on non-significant results?
I'm not sure where you're coming from on this issue. ... What, in fact, is your deal?
I avidly opposed to theocracy as a form of government, a fortiori a nuclear armed theocracy. I am opposed to the Iranian regime even though I don't live there (some might point out it is none of my business). I don't consider A'jad to be entirely a sane man (though by the standards of his culture he may well be). I would dearly have loved to see him defeated (and he may still be, extra-democratically).
However I also endorse an evidence based view of reality, and I don't think it is valid to construe the world in a way which lacks evidence merely on the basis that I might like it to be that way. I believe that, unpalatable though we may find it, we might have to accept that a majority of Iranians disagree with us. Moreover, in the absence of clear evidence to the contrary (eyewitness, statistical, or whatever), I strongly believe the only (rebuttable) presumption we are entitled to draw (upon evidence-based criteria) is that the election result reflects the will of the majority of Iranian voters. To draw any other conclusion in the absence of compelling evidence to the contrary is, IMHO, to fall victim to the trap of believing in one's own propaganda.
Better to be despised for too anxious apprehensions, than ruined by too confident a security. --Edmund Burke